Chapter 20 – Six Sigma

Management
 “Six Sigma” (a registered trademark of the Motorola
Corp.) is the relentless and rigorous pursuit of the
reduction of variation in all critical processes to
achieve continuous and breakthrough improvements
that impact the bottom line of the organization and
increase customer satisfaction.
 It is an organizational initiative designed to create
manufacturing, service and administrative processes
that produce approximately 3.4 defects per million.
Opportunities (DPMO).
 Relationship between the Voice of the
Customer and the Voice of the Process
 Six Sigma management promotes the idea that the
distribution of output for a process (the Voice of the
Process) should take up no more than half of the
tolerance allowed by the specification limits (the
Voice of the Customer)
 This assumes that the measurement data is from a
stable and normal distribution of output whose mean
can shift by as much as 1.5 standard deviations over
time.
 The following figure shows the Voice of the Customer
as spoken in the language of a nominal value, m, and
the lower and upper specification limits, LSL and
USL, for a quality characteristic.
 This assumes
– Distribution of process output is measurement data
– The process is stable
– The distribution is normal
– The process average is on nominal
– The distance between nominal and either specification limit
is 3 times the process standard deviation
 If these five conditions are met we all the process a
3-sigma process.
 A 3-sigma process will produce 2,700 defects per
million opportunities.
LSL
3-sigma Process USL
 An Example of a 3-sigma Process
 The number of days to complete a month accounting
report is stable and normally distributed with an
average of 7 days and a standard deviation of 1 day.
 The lower specification limit is 4 days and the upper
specification limits is 10 days.
 Next we see a quality characteristic represented by
measurement data that:
– Is stable
– Is normally distributed
– Has a process output that can shift by as much as 1.5
standard deviations on either side of nominal
– Has a distance between nominal and either specification
limit of 3 standard deviations of process output.
 This is a 3-sigma process with a 1.5 sigma shift in the
mean.
 This process will generate 66,811 defects per million
opportunities or 93.33189% of its output between the
specification limits.
Accounting process with process average shifts of 1.5
standard deviations.
 Next we see a scenario where the voice of the
process takes up only half the distance between the
specification limits. The process mean remains the
same, but the process standard deviation has been
reduced to one half-day
Output for the monthly accounting process with the
standard deviation reduced to one-half day through process
improvement activities.
 Next we will see the voice of the customer and the
voice of the process for a quality characteristic that:
– Is represented by measurement data
– Is stable
– Is normally distributed
– Has a process average that can shift by as much as 1.5
standard deviations on either side of nominal
– Has the distance between nominal and either specification
limit is 6 standard deviations of the process output.
 This is a process that will generate 3.4 defects per
million opportunities, or 99.99966% of its output
within specification limits in the near future.
Six sigma accounting process with a 1.5 sigma shift in
the mean
 The DMAIC Model

 The model that is used to improve a process in Six

Sigma management is called the DMAIC model. This
stands for:
– Define
– Measure
– Analyze
– Improve
– Control
 Define Phase
 The define phase involves:
– Preparing a business charter (rationale for the project)
– SIPOC Understanding the relationships between
• Suppliers
• Inputs
• Process
• Outputs
• Customers
– Gathering and analyzing voice of the customer data to
identify the critical-to-quality (CTQ) characteristics
important to customers
– Developing a project charter (statement of the project)
 Measure Phase
 The measure phase involves
– Developing operational definitions for each CTQ
characteristic
– Performing studies to determine the validity (repeatability
and reproducibility) of the measurement procedure for each
CTQ
– Collecting baseline capabilities for each CTQ
– Determining the process capability for each CTQ
 Analyze Phase
 The analyze phase involves
– Identifying upstream variables (x’s) for each CTQ
– Operationally defining each x
– Collecting baseline data for each x
– Performing studies to determine the validity (repeatability
and reproducibility) of the measurement process for each x
– Establishing baseline capabilities for each x
– Understanding the effect of each x on each CTQ
 Improve Phase

 The improve phase involves

– Designing experiments to understand the relationships
between the CTQ’s and the x’s
– Determining the optimal levels of critical x’s that optimize the
spread, shape and center of the CTQ’s
– Developing action plans to implement the optimal level of the
x’s into the process under study
– Conducting a pilot test of the revised process
 Control Phase

 The control phase involves

– Avoiding potential problems with the revised settings of the
x’s through risk abatement planning and mistake-proofing
– Standardizing successful process revisions in training
manuals
– Controlling the revised settings of the critical x’s
– Turning the revised process over to the process owner for
continuous improvement using the PDSA cycle
 Benefits and Costs of Six Sigma
Management
 A successful six sigma program will yield the
following benefits to the management of the
organization
– Improved process flows
– Improved communication through six sigma terminology (for
example, DPMO and process sigma)
– Reduced cycle times
– Enhanced knowledge and enhanced ability to manage that
knowledge
– Higher levels of customer satisfaction
– Higher levels of employee satisfaction
– Increased Productivity
– Reduced total defects
– Decreased work-in-progress (WIP)
– Decreased inventory
– Improved capacity and output
– Increased quality and reliability
– Decreased unit costs
– Increased price flexibility
– Decreased time to market
– Faster delivery time
– Increased liquid capital
 Resources required by Six Sigma management
– Training time costs
– Material costs
– Training manual development costs
– Administrative and operating costs for DMAIC projects
– Infrastructure costs such as the sots of constructing and
using organizational metric tracking systems
– Monitoring DMAIC project costs
Anecdotal evidence strongly indicates that he benefits
of a Six Sigma process far outweigh the costs.
Six Sigma Roles and Responsibilities

 The Senior Executive provides the impetus,

direction and alignment necessary for the ultimate
success of Six Sigma management.
 Roles and responsibilities of a senior executive
– Leads the Executive Committee (EC)
– Participates in high-level policy management and cross-
functional project teams.
– Monitors and balances all Six Sigma activities to avoid local
optimization and organizational suboptimization.
– Maintains a long-term view of the organization and acts as a
liaison to stakeholders
– Conducts Presidential reviews of Six Sigma daily
management and cross functional management projects.
 Top management must be totally committed to the
process.
 The CEO must allocate substantial resources to Six
Sigma management.
 Leadership must be willing to engage all levels of the
organization through organizational metric tracking
systems and daily management and cross-functional
Six Sigma and DMAIC projects.
 The CEO must have a “burning desire” to create
revolutionary change in the enterprise.
 Members of the Executive Committee (EC) must
operate with the same level of commitment as the
senior executive.
 The EC must
– Create an organizational metric tracking system
that cascades, or deploys, key objectives and key
indicators throughout the organization via policy
management
– Empower the Policy Deployment Committee
(PDC) to deploy key objective throughout the
organization through policy management.
– Manage the organizational Six Sigma project
portfolio toward optimization of the entire
organization’s bottom line
– Use catchball to establish a reasonable and
equitable project portfolio for each
division/department within the organization.
– Improve the Six Sigma process through constant
iteration of the PDSA cycle.
– Provide the resources necessary for Six Sigma
management.
 The role and responsibilities of a black belt are:
– Full time change agent
– Master of the change process rather than a master of the
process under study
– Supervise green belts working on a Six Sigma project
– Prepare a draft project charter for the Six Sigma projects
under his/her supervision.
– Work closely with a project team to keep it functioning and
progressing toward a conclusion to its Six Sigma project.
– Communicate with the individual responsible for the financial
and political well-being of the team
– Serve as the team leader for Six Sigma projects
– Help team members analyze data and design experiments
 – Provide training in Six Sigma theory, tools and methods
– Help team members prepare for management and
presidential reviews
– Recommend Six Sigma teams for six Sigma projects
– Lead and coach green belts leading simpler Six Sigma
projects
 A black belt must pass a certification examination and
lead at least two successful Six Sigma projects.
 Most green belts serve as team members on Six
Sigma projects.
 If a green belt acts as a team leader for simpler
projects he/she:
– Prepares a draft charter for the Six Sigma project
– Selects the project team members
– Communicates with the champion, black belt and process
owner concerning the status of the project
– Facilitates the project members
– Provides training in the basic Six Sigma tools and methods.
 A green belt must pass a certification examination
and participate in at least one successful Six Sigma
project.
 A master black belt’s responsibilities are:
– To be a proven team leader and technical expert
– To be a teacher and mentor of black belts and green belts
– Simultaneously supervise several black belts and green
belts
– To be an ambassador (in conjunction with the senior
executive, members of the executive committee, and
champions) of Six Sigma management
– To work continuously to improve and innovate the Six Sigma
management process.
 A master black belt must complete a certification
examination and successfully supervise at least two
black belts in their completion of two successful Six
Sigma projects each.
 A champion is a member of the EC, or at least an
individual who is trusted to report directly to a
member of the EC.
 A champion takes a very active leadership and
sponsorship role in implementing Six Sigma
management, and works closely with the EC, black
belts and master black belts to:
– Translate key objectives and key indicators from his section
of the organizational metric tracking system into Six Sigma
projects as a part of policy management
– Prepares an initial draft of a project charter for each Six
Sigma project under his auspices
– Assigns green belts and black belts to the Six Sigma
projects under his auspices
– Removes obstacles to the effective functioning of the Six
Sigma project teams under his auspices
 – Provides a direct line of communication between a Six sigma
project team and the EC
– Obtains and manages the resources necessary for a Six
Sigma project team to meet its project charter
– Conducts management review of each Six Sigma project
team under his auspices focused by providing direction and
guidance on the attainment of its project charter

A champion must pass a certification examination.

 A process owner is the individual who has the
ultimate authority to change a process.
 The process owner should be identified for every
project or task that is entered onto an organizational
metric tracking system.
 The process owner:
– Monitors the performance of his/her process through key
indicators
– Empowers the people who work in the process
– Works with all Six Sigma project teams in the area to enable
them to successfully complete their projects
– Manages the process after completion to the Six Sigma
project to sustain the gains made.
 – Continues to improve and/or innovate the process through
the application of the PDCA cycle.

 A process owner should pass the champion

certification examination.
Six Sigma Terminology

 Unit – A unit is the item (e.g. product or component,

service or service step, or time period) to be studied
with a Six Sigma project.
 Defective – A nonconforming unit is a defective.
 Defect – A defect is a nonconformance on one of
many possible quality characteristics of a unit that
causes customer dissatisfaction.
 Defect Opportunity – a defect opportunity is the
most fundamental area for a defect. There may be
several opportunities for defects within a defined unit.
 Defects per Unit (DPU) - Defects per unit refers to
the average of all defects for a given number of units,
that is, the total number of defects for n units divided
by n.
 Defects per Opportunity (DPO) – Defects per
opportunity refers to the number of defects divided by
the number of opportunities.
 Defects per Million Opportunities (DPMO) –
DPMO equals DPO multiplied by 1,000,000.
 Observed Yield – Observed yield is the proportion of
units within specification divided by the total number
of units.
 Rolled Throughput Yield (RTY) – RTY is the
product of the observed yield s from each step in a
process. It is the probability of a unit passing through
all steps of a process and incurring no defects.
 Process Sigma – Process sigma is a measure of the
process performance determined by using DPMO
and a normal distribution table. Process sigma is a
metric that allows for process performance
comparisons across processes, departments,
divisions, companies, and countries. In Six Sigma
terminology, the sigma value of a process is a metric
used to indicate the number of defects per million
opportunities, or how well the process is performing
with respect to customer needs and wants.
 A Six Sigma Case Study
Background
 Paper Organizers International (POI) offers a full
range of filling, organizing, and paper shuffling
services.
 POI purchases Metallic Securing Devices (MSDs),
staplers, hole punchers, folders, three-ring binders,
and a full range of related products to serve its
customers in paper handling needs.
 The employees, or internal customers, of Paper
Organizers use MSDs to organize piles of paper
pending placement into folders or binders.
 The purchasing department has noticed an
increase in complaints from employees in the
Paper Shuffling Department (PSD) about MSDs
breaking and failing to keep papers together.
 This creates opportunities for client papers to be
mixed together.
 The Purchasing Department would like to improve
the purchasing process for purchasing MSDs to
eliminate complaints from employees in the Paper
Shuffling Department.
 Origin of the MSD Six Sigma Project

Mission Statement:
 Put the Right Information in the right Place.
 RIP it!
POI’s Business Objectives and Indicators with Potential Six Sigma Projects

President Director of Paper Shuffling Department

Business Business Area Objectives Area Indicators Potential Six

Objectives Indicators Sigma Projects

Increase the # orders /month Increase the # orders in New Customer

number of orders (c-chart) number of orders PSD/month Promotions
in PSD (c-chart) Project

Increase the 1. Average # of Increase the 1. Average # of Existing Customer

number of POI services utilized number of services utilized Promotions
services (filing, per customer / services utilized per PSD Project
organizing, etc) quarter by each customer customer/quarter
utilized by each 2. Standard in PSD 2. Standard
customer deviation of deviation of
number of number of
services utilized services utilized
per per PSD
customer/quarter customer/quarter
(x-bar and s chart) (x-bar and s chart)

Minimize Production Minimize Production costs MSD Quality

production costs costs/month production costs in PSD/month Project
(I and MR chart) in PSD (I&MR chart)

Eliminate # of employee Eliminate # of employee Employee Morale

employee complaints/month employee complaints from Project
complaints (c-chart) complaints from PSD/month
PSD (c-chart)
 Individual and Moving Range Chart of Monthly
Production Costs in the Paper Shuffling Department
I and MR Chart for Production C

1400000 UCL=1393879
1300000

Individual Value
1200000
1100000 Mean=1096880
1000000
900000
800000 LCL=799881

Subgroup 0 10 20 30 40 50 60 70

400000
UCL=364863
300000
Moving Range

200000

100000 R=111672

0 LCL=0

Production costs are stable (no special causes such as points

beyond a control limit or too many runs up and down, etc.) in
the PSD with an average monthly cost of $1,096,880 and a
standard deviation of R / d  $111,672/1 .128  $116,672 2
Distribution of Monthly Production Costs in the PSD

Histogram of Production Costs in PSD

10

Fr equen cy
5

0
890000 930000 970000 1010000 1050000 1090000 1130000 1170000 1210000 1250000 1290000

Production Costs in PSD

•Production costs are approximately normally distributed.

•Team members discovered that PSD management considers
monthly production costs to be very high given the volume of
work being processed by the department.
 Prioritization of Six Sigma Projects
 Top management generated four potential Six
Sigma Projects.
 The following table is used to estimate the
importance of each potential Six Sigma project on
POI’s business objectives.
  Potential Six Sigma Projects
 
 
  New Existing MSD Employee
Business Objectives Customer Customer Quality Morale
Promotions Promotions Project Project
Project Project

  Increase W .35 3 3 0 0
the number E
of orders I
G
H
Increase T .10 1 3 0 0
the number S
of POI
services
utilized by
each
customer
  Minimize   .40 0 0 9 3
production
costs

  Eliminate   .15 0 0 9 9
employee
complaints

Weighted Average of Potential Six 1.15 1.35 4.95 2.55

Sigma Projects
 The cell values are assigned by top management and are
defined as follows: 0 = no relationship, 1 = weak
relationship, 3 = moderate relationship, and 9 = strong
relationship.
 The Finance Department developed the importance
weights for each business objective to maximize the
impact of Six Sigma projects on the bottom-line of the
organization.
 Consequently, the most critical project in respect to the
business objectives is the MSD quality project, see 4.95 in
the last row of the table.
 The Champion and Process Owner of the “MSD process”
prepared an initial project charter. It presented the
business case for the MSD quality project to the members
of the MSD quality project team.
 Define Phase
 The define phase has three components:
– prepare a business case with a project charter
– do a SIPOC analysis
– conduct a “Voice of the Customer” analysis.

Prepare a Business Case with a Project Charter

 Preparing a business case with project charter

requires team members to answer the following
partially redundant questions. The redundancy in
the questions helps team members distill the
critical elements of the business case.
1. What is the name of the process?
MSD Purchasing Process.
– The first step in the supply chain for the “MSD process”
is the process for purchasing MSDs.
– Hence, the first operation to be investigated by MSD
quality project team members is the process for
purchasing MSDs.
– Team members may study other factors that affect the
quality of MSDs such as method of use or shelf life at a
later time.
2. What is the aim of the process?
 The aim of the purchasing process as it relates to this project
is to purchase MSDs that improve the productivity and
morale of the employees in the Paper Shuffling Department
(PSD).

3. What is the business case (economic rationale) for the

project?

 Question 3 is answered by addressing the following sub-

questions.
3a. Why do the MSD project at all?
 According to a judgment sample of 3 employees and 2
managers from the PSD, team members determined that
MSDs that cannot withstand 4 or more bends are
unacceptable because they are unlikely to remain intact
throughout the paper shuffling processes and will not hold
papers tightly; this is called durability.
 Defective MSDs create costs for POI, for example:
– papers from different clients may get mixed together if
not properly bound requiring additional processing time
– employees may have to use multiple MSDs for one
project creating additional material costs
– employees get frustrated and do not perform their jobs
efficiently and productively increasing labor costs.
 Additionally, team members discovered that a
large proportion of the boxes containing MSDs
arrive to the PSD with 5 or more broken MSDs;
this is called functionality.
 This creates additional processing costs for POI,
for example
– increased unit costs
– frustrated and non-productive employees and
managers.
 Team members used the same judgment sample
as above and determined that approximately 60%
of individual MSDs do not meet durability criteria
and 60% of MSD boxes do not meet functionality
criteria. They used the following survey
questionnaire and the data matrix.
 Survey
Name: ____________________________
1. Please estimate the percentage of MSDs that cannot withstand 4 or more bends. _____
2. Please estimate the percentage of MSD boxes that contain greater than 5 broken MSDs.
_____

Survey Number Response Q1 Response Q2

1 55 70

2 50 55

3 60 65

4 65 60

5 70 50

Average 60 60
3b. Why do the MSD project now?

 The Paper Shuffling Department is experiencing very high monthly production costs.
 Also, internal customers, including managers and hourly employees, are submitting an
increased number of complaints as recorded in the Purchasing Department’s complaint
log for the fiscal year 2000:
– 14 in the first quarter
– 18 in the second quarter
– 32 in the third quarter

 There are 100 hourly workers in the Paper Shuffling Department.

3c. What business objectives are supported by the MSD
project?

 The MSD project is most strongly related to the

“minimize production costs” and “eliminate employee
complaints” business objectives.
3d. What are the consequences of not doing the project?
 The consequences of not doing the project are decreased
profit margins due to higher production costs and increased
employee complaints due to frustration with materials.

3e. What projects have higher or equal priority?

 At this time, the MSD quality project has the highest priority.
4. What is the problem statement? What is the pain?
 Low quality MSDs create additional production costs
and employee frustration.

5. What is the goal (desired state) for this project?

 The Champion and Process Owner of the MSD
process initially determined that a 100-fold
improvement in MSD quality (durability and
functionality) should be the goal for the Six Sigma
project.
 They derived the concept of a 100-fold improvement
from Motorola’s 1986 stated improvement rate of
10-fold every 2 years, or a 100-fold every 4 years
during the kickoff of the Six Sigma effort.
 Since 100-fold improvement means the DPMO would
decrease from 600,000 to 6,000, and a DPMO of
6210 represents a 4-sigma process, team members
decided to use 4-sigma as the goal for the MSD
project.
6. What is the project scope?
 Question 6 is answered by answering the following sub-questions.
6a. What are the process boundaries?
 The starting point for the project is when the Purchasing Department
receives purchase orders from the PSD.
 The stopping point for the project is when the PSD places MSDs into
inventory.
6b. What, if anything, is out-of-bounds?
 The project team cannot change the way employees handle or use
MSDs.
6c. What resources are available for the project?
 The budget for the MSD project is $30,000.00. This includes
estimated hourly salaries of project participants.
 Team members, Brian Mercurio and Jeremy Pressman, are the
only project participants that will incur additional job
responsibilities as result of the project.
 Budget estimates show “opportunity cost” and “hard costs.” The
estimated hard costs ($10,500) and total costs ($26,040) are
less than the budget of $30,000.
NAME POSITION Estimated Expected Expected Expected
Salary / Number of Opportunity Hard
Hour Hours Costs for 21 Costs for
Per Week weeks 21 weeks
(direct
labor
costs)

Howard Gitlow Champion $100 2 $4,200  

Dana Rasis Process Owner $50 2 $2,100  

Bettina Arguelles Black Belt $50 5 $5,250  

Brian Mercurio Team Member $25 10 $0 $5,250

Jeremy Pressman Team Member $25 10 $0 $5,250

Lindsey Barton Finance Rep. $45 2 $1,890  

Mary Montano IT Rep. $50 2 $2,100  

TOTAL       $15,540 $10,500

6d. Who can approve expenditures?
 Only the Process Owner, Dana Rasis, can approve expenditures.

6e. How much can the team spend beyond $30,000.00 without seeking
additional authority?
 Nothing.

6f. What are the obstacles and constraints of the project?

 The team must work within a $30,000 budget and a 21week time constraint.
6g. What time commitment is expected of team members?
 Team members are expected to be present at weekly
Friday morning meetings from 8:00 a.m. until 9:00 a.m.
 Team members are also expected to provide progress
of project tasks at each meeting. Completion of project
tasks may require additional hours of work per week.

6h. What will happen to each team member’s regular job

while he or she is working on the project?
 If any, overtime hours will be compensated for team
members and support staff. Note: The estimated rate
for overtime labor is 1.5 times normal labor. Overtime
labor is not included in the budget.
6i. Is there a Gantt chart for the project?
 Yes.
Steps Resp. Week

1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 1 1 1 1 2 2
0 1 2 3 4 5 6 7 8 9 0 1
Define BA X X X X X X

Mesure BA X X

Analyze BA X X X

Improve BA X X X X X X

Control BA X X X X
7. What are the benefits of the project?
 The soft benefits of the project include eliminating
complaints from the Paper Shuffling Department
and increasing employee morale.
 The hard (financial) benefits of the project are
minimizing labor costs and material costs. The
hard cost benefits are estimated from the following:
100 employees in the Paper Shuffling Department
x 40 hours/week/paper shuffling employee
x 10% of time devoted to clipping
@ 400 hours/week devoted to clipping in PSD
x $25/hour/paper shuffling employee
$10,000/week devoted to clipping
x 50 weeks/year
$500,000/year devoted to clipping
x .60 defective clips (judgment sample estimate of durability of the
current system) Broken clips are not selected for use on jobs. This
makes 0.6 a conservative estimate of the percentage of defective clips in
the current system. Note: This conservative estimate does not include
problems arising from defective clips not detected until after they have
been used and have caused failure on the job.
$300,000/year on defective clipping for current system
 
x .0062 defective clips (durability of the proposed system) Again, broken
clips are not selected for use on jobs.
$3100/year on defective clipping for proposed system
 Hence, the annual savings on labor costs from
improving MSD purchasing process is $296,900
($300,000 - $3,100).
 The PSD incurs a 10% annual employee turn over.
 To capitalize on savings in labor costs, the
department will now higher 4 new employees
instead of 10 new employees, for a savings of 6
full-time employees ($296,900/$25 = 11,876 hours;
11,876/40 hours per week/50 weeks per year =
5.938= ~ 6 employees saved).

Note: Alternatively, the PSD may now serve more

customers with their current employee base.
 Material Costs:
100 employees in the Paper Shuffling Department

x60 projects/week/paper shuffling employee

x50 weeks/year

@ 300,000 projects/year requiring 3,000,000 MSDs (10 clips per project on average)

x.60 defective clips (judgment sample estimate of current system).

7,500,000* clips must be used to complete 300,000 projects

x.01/clip

@ $75,000/year on clips in current system

x.0062 defective clips (proposed system)

3,018,000** clips must be used to complete 300,000 projects

x.01/clip

@ $30,180/year on clips in proposed system

 Material Costs:
Note: * 1/(1-.6) = 2.5 clips needed to get a good
clip. So, 3,000,000 x 2.5 = 7,500,000
Note: ** 1/(1-.0062) = 1.006 clips needed to get a
good clip. So, 3,000,000 x 1.006 = 3,018,000.
 
 Hence, the annual savings on material costs from
improving MSD purchasing process is $44,820
($75,000 - $30,180).
 This yields an annual total hard benefit savings of
$341,720.00.
8. What are the roles and responsibilities of team members?

Project Name: MSD Purchasing Process

Role Responsibility Stakeholder Supervisor’s

Signature
Signature Date

Champion Howard Gitlow HG 9/1/2000  

Process Owner Dana Rasis DR 9/1/2000  

Team Leader Bettina Arguelles BA 9/2/2000  

Team Member 1 Bryan Mercurio BM 9/3/2000  

Team Member 2 Jeremy Pressmen JP 9/3/2000  

Finance Rep Lindsey Barton LB 9/4/2000  

IT Rep Michelle Montano MM 9/4/2000  

Do a SIPOC Analysis
 The second part of the define phase requires that
team members perform a SIPOC analysis.
 A SIPOC analysis is a simple tool for identifying
the Suppliers and their Inputs into a Process, the
high level steps of a process, the Outputs of the
process, and the Customers segments interested
in the outputs.
 Start

Purchasing
receives order from
Paper Shuffling
Department

Purchasing agent
calls vendor
No

Does vendor
have MSD
in stock?

Yes
Place order with
vendor

Receive order from

vendor

Store product
received into
inventory (new
boxes go on bottom
back of shelf)

PSD removes
products from
inventory

PSD uses Product

Stop
Conduct a “Voice of the Customer” Analysis
 The third part of the define phase involves team
members collecting and analyzing “Voice of the
Customer” data. “Voice of the Customer” data is
verbal or written information collected from a
sample of users, in a selected market segment.
 The questionnaire used to collect data from users
of MSDs in the PSD is:
Questions

What emotions come to mind when you think about MSDs?

What needs and wants come to mind when you think about MSDs?

What complaints or problems would you like to mention about MSDs?

Note: These questions do not consider the opinions, feelings and attitudes of the
upstream, downstream, and external customers of the PSD.
 Team members analyze the “Voice of the
Customer” data by market segment.
 Next, they use all the raw “Voice of the Customer”
data points to create affinity diagram.
 Themes, called focus points using numbers linking
columns 2 and 3 in column 3 in the following table
are identified.
 Next, team members identify the engineering issue
underlying each focus point, called cognitive
issues.
 Then, team members convert each cognitive issue
into one or more quantitative engineering
variables, called Critical-To-Quality (CTQ)
variables.
 Finally, team members develop technical
specifications for each CTQ.
 Affinity Engineering
Selected Diagram issue
Market Raw “Voice of the Customer” theme (Focus (Cognitive Tech.
Segment data Point) issue) CTQ Spec.
Paper “My employees are frustrated Variation in Durability Ability to >= 4 bends
organizing about the MSDs. They complain durability (1) withstand without
managers that they break too fast” 1&2 bending breaking
“My employees are complaining Variation in Color The = 1 color of
that the MSDs are not holding color (2) number of MSDs
up during the organizing different
process.” 1 MSD colors
“The employees are Variation in Functionality The <=5
complaining that the color of the functionality number of broken
MSDs change from one day to (3) broken MSDs in a
the next. It seems to be MSDs in a box
confusing them.” 2 box
“My employees are very
unhappy with the purple and
blue MSDs. They would prefer
only one color of MSDs be used
consistently” 2
“My employees say that more
than 5 MSDs per box arrive
broken.” 3
“I’ve heard from numerous
employees that the MSDs
coming straight from inventory
are already broken.” 3
...
 Affinity Driving
Selected Diagram issue
Market Raw “Voice of the Customer” theme (Focus (Cognitive Tech.
Segment data Point) issue) CTQ Spec.
Hourly "The MSDs are falling apart Variation in Durability Ability to >= 4 bends
employees before we are ready to file the durability (1) withstand without
papers in to binders. An MSD bending breaking
should be able to take at least 4
bends." 1
"The MSDs aren't helping us to Variation in Color The = 1 color of
do our work efficiently." 1 & 2 color (2) number of MSDs
different
MSD colors
“I would prefer if we only used Variation in Functionality The <=5
one color of MSDs.” 2 functionality number of broken
(3) broken MSDs in a
MSDs in a box
box
“I don’t understand why we use
different colors of MSDs.” 2
"The MSDs just break when
trying to bend them over the
paper stacks. They should take
at least 4 bends" 1
“It is very frustrating when you
open a brand new box of MSDs
and find that more than 5 of the
clips are already broken.” 3
“It is very time consuming to sift
out the broken MSDs from a
brand new box coming straight
from inventory.” 3
...
 Returning to the first part of the define phase, team members can now define the project’s objectives.

• Project Objective 1: Decrease (direction) the percentage of MSDs that can not withstand 4 or more bends without
breaking (measure) bought by the Purchasing Department (process) to 00.62% (goal) by January 1, 2003 (deadline).
Go for 4-sigma!

• Project Objective 2: Decrease (direction) the percentage of boxes of MSDs with more than 5 broken clips (measure)
bought by the Purchasing Department (process) to 00.62% (goal) by January 1, 2003 (deadline). Go for 4-sigma!
 
• A correlation exists between the project objectives. A broken MSD cannot withstand 4 or more bends because it is
already broken. Improving the percentage of functional MSDs per box will increase the percentage of MSDs that can
withstand 4 or more bends.
 Measure Phase
• The measure phase has three steps:
• operationally define the CTQs
• perform a gage R&R study on each CTQ
• develop a baseline for each CTQ.

• Operationally define the CTQs

• First, team members operationally define durability and functionality by:

• establishing criteria for durability and functionality
• developing a test for each set of criteria
• formulating a decision rule for each criteria.
Criteria for a selected MSD are:
 Test for a selected MSD

1. Select the “top-front” box of MSDs on the shelf in the

inventory room.
2. Close your eyes, then open the box of MSDs, then
haphazardly select one intact MSD. No switching is
allowed.
3. Utilize the criteria for the selected MSD
4. Count the number of bends until breaking

Decision for a selected MSD

If the number of bends is  4, then MSD is conforming.

If the number of bends is < 4, then MSD is defective.
Operational Definition for CTQ2: Functionality
Criteria for a box of MSDs: Count the number of “broken” clips. A
clip is “broken” if it is in 2 pieces, regardless of the relative sizes of
the pieces. Clips can be broken only into two pieces.
Test for a box of MSDs:
1. Select the “top-front” box of MSDs on the shelf in the inventory
room.
2. Count the number of “broken” clips.

Decision for a box of MSDs

If the number of MSDs that are broken  5, then the box of MSDs is
conforming.
If the number of MSDs that are broken > 5, then the box of MSDs is
defective.
 
The same box of MSDs is used for both operational definitions.
Perform a gage R&R study on each CTQ.
•Team members conduct an attribute Gage R&R
(Repeatability and Reproducibility) study on the
measurement system of each CTQ to determine if it is
adequate for the needs of the project.

•The measurement of durability requires a destructive

test, hence, a simple Gage R&R study was not done for
durability at this time.

•In the near future, an operational definition of the testing

process for durability will be established and testing will
be audited to assure consistency.
• The measurement system for functionality is studied using the
following sampling plan.

1. A shelf in the storage area contains boxes of MSDs purchased

throughout the week. There are different types of MSD boxes in
the storage area (different vendors, sizes, etc.).
2. The Gage R&R study required 2 inspectors to sample the same
10 boxes of MSDs twice.
3. The top 10 boxes on the front of the shelf were selected for the
Gage R&R study.
4. The study is repeated as is deemed necessary by PSD
management.

• Two PSD managers have the responsibility of inspecting the

MSDs for functionality; they are called Inspector 1 (Tom) and
Inspector 2 (Jerry). Both Tom and Jerry counted the number of
defective MSDs, twice, in random order.
 Inspecto Inspecto
Box Count Functionality Box Count Functionality
r r
1 1 1 10 6 1 1 9
1 1 2 10 6 1 2 9
1 2 1 10 6 2 1 9
1 2 2 10 6 2 2 9
2 1 1 9 7 1 1 6
2 1 2 9 7 1 2 6
2 2 1 9 7 2 1 6
2 2 2 9 7 2 2 6
3 1 1 5 8 1 1 6
3 1 2 5 8 1 2 6
3 2 1 5 8 2 1 6
3 2 2 5 8 2 2 6
4 1 1 4 9 1 1 9
4 1 2 4 9 1 2 9
4 2 1 4 9 2 1 9
4 2 2 4 9 2 2 9
5 1 1 5 10 1 1 11
5 1 2 5 10 1 2 11
5 2 1 5 10 2 1 11
5 2 2 5 10 2 2 11
 Gage name:
Date of study:
Runchart of Fuctionality by Box, Inspector Reported by:
Tolerance:
Misc:

11 1
10 2
9

Fuctionality
8
7
6
5
4
Box 1 2 3 4 5

11
10
9
Fuctionality

8
7
6
5
4
Box 6 7 8 9 10

A Gage Run chart shows that there is no variation within inspectors or

between inspectors. All the variation is between the 10 boxes of
MSDs. Hence, the measurement system is acceptable to measure
functionality.
Develop a baseline for each CTQ.
 At the beginning of each hour, one box of MSDs is selected from the storage area.

 The procedure for selecting a box of MSDs is simply to select the top-front most box on the shelf. The selection process was not altered during a sampling period of two 8-hour
shifts.

Baseline capability data is as follows.

Hour Durability Functionality
Shift 1 – Hour 1 5 12
Shift 1 – Hour 2 7 4
Shift 1 – Hour 3 3 8
Shift 1 – Hour 4 2 6
Shift 1 – Hour 5 9 1
Shift 1 – Hour 6 2 5
Shift 1 – Hour 7 1 11
Shift 1 – Hour 8 1 9
Shift 2 – Hour 1 12 6
Shift 2 – Hour 2 9 6
Shift 2 – Hour 3 3 9
Shift 2 – Hour 4 1 5
Shift 2 – Hour 5 1 4
Shift 2 – Hour 6 1 5
Shift 2 – Hour 7 1 9
Shift 2 – Hour 8 4 10
Yield 6/16 = .375 6/16 = .375
• The yield for durability and functionality are both 0.375 as
determined by the number of tests out of 16 trials shown in the
previous table that met their respective CTQ’s (i.e., at least 4
bends for durability, no more than 5 broken MSD’s per box for
functionality).
• This indicates very poor levels of durability and functionality
for the MSDs received into the Paper Shuffling department and
supports the initial yield estimates of 40.0%, or 60% defective
MSDs.
 
• An individuals and moving range (I-MR) chart for the
durability baseline data indicates that the variability of
“durability” is not stable over time, see bottom panel of the
following chart.
• An investigation of the range between the eight and ninth
MSDs did not reveal any obvious special cause of variation
that could be used to improve the durability of MSDs.
 I and MR Chart for Durability

15

Individual Value
UCL=12.21
10

5
Mean=3.875

-5 LCL=-4.458

Subgroup 0 5 10 15

1
10 UCL=10.24
Moving Range

5
R=3.133

0 LCL=0

Note: Use of the I-MR chart assumes approximate normality of the CTQ (durability). The
durability data is clearly not normally distributed, see dot plot that follows.
• Hence, use of the durability I-MR chart is not advised
at this time.
• However, the distribution of durability may
approximate a Poisson distribution.
• Consequently, team members constructed a c-chart
for the “count of bends” before each MSD breaks,
see next slide.

• Note: If the “durability” were measured using a continuous

measurement system allowing for fractional number of bends before
breaking, then a log or similar transformation of the distribution may be
appropriate before using an Individuals-Moving Range charting
procedure.
The c chart indicated a possible special cause during Shift 2 - Hour 1 when 12
bends was observed for the durability test. Further investigation and notes related
to the test did not reveal any obvious differences between the MSD tested and the
others, although during the first hour the tester indicated that he may have bent the
MSD slower than usual during the test which may have caused less stress and
consequently more bends.
A c-chart for “functionality” indicates that is stable over
time.
The functionality data appears to be approximately Poisson
distributed (due to a Goodness of Fit test). Hence, use of
the functionality c-chart is acceptable at this time.
Finally, team members estimate the current process
performance for each CTQ in the following table.

CTQs Yield DPMO  Process Sigma

  Current Desired Current Desired Current Desired

Durability 37.50% 99.38% 625,000 6210 ~1.2 4.0

Functionality 37.50% 99.38% 625,000 6210 ~1.2 4.0

Notice the desired 100-fold improvement shown in the DPMO

columns (Current = 625,000 and Desired = 6,210). This is
consistent with the goals stated in question 5 of the Define Phase
of the DMAIC model.
 Analyze Phase
• The Analyze phase has five steps:

1. Develop a more detailed process map (that is, more detailed than the
process map developed in the SIPOC analysis of the Define phase)
2. Construct operational definitions for each input or process variable
(called Xs)
3. Perform a gage R&R study on each X (test the adequacy of the
measurement system)
4. Develop a baseline for each X
5. Develop hypotheses between the Xs and Ys. The Ys are the output
measures used to determine if the CTQs are met.

• First, team members prepare a detailed process map identifying and

linking the Xs and Ys.
 Start

Purchasing
receives order from
Paper Shuffling
Department

Purchasing agent
calls vendor
No

Does vendor
have MSD
in stock?

Yes
Place order with
vendor X1 – Vendor (Ibix or Office Optimum)
X2 – Size (Small or Large)
Receive order from X3 – Ridges (With or Without)
vendor X4 = Cycle time from order to receipt for MSDs
X5 = Discrepancy in count from order placed and
MSDs placed into
order received
inventory (new boxes
go on the bottom
back of shelf) X6 = Cycle time to place product in inventory

PSD removes
box from X7 = Inventory shelf time (in days)
inventory

PSD uses MSDs X8 = Type of usage (Large stack of paper or

Small stack of paper)

Stop
Develop an operational definition for each X.
•Team members develop an operational definition for each X
variable identified on the process map. The operational
definitions for X1, X2, X3 and X8 relate to individual MSDs and
are shown below.

Criteria: Each X conforms to either one or the other of the options.

X1 Vendor Ibix Office Optimum

X2 Size Small (stock size) Large (stock size)

X3 Ridges With ridges Without ridges

X8 Type of Usage Large stack of paper Small stack of

(# papers is 10 or more) paper
(# papers is 9 or
less)
Test: Select MSD.
Decision: Determine X1, X2, X3, and X8 options for the selected
MSD.
The operational definitions for the procedures used to measure X 4, X5, X6 and X7 are:

Criteria: Compute cycle time in days by subtracting the order date from the date on the bill of lading.

Test: Select a box of MSDs upon receipt of shipment from vendor. Compute the cycle time.
Decision: Determine X4 for the selected box of MSDs.

X4 Cycle time from order to In days

receipt for MSDs
Criteria: Count the number of boxes of MSD received for a given order. Subtract the
number of boxes ordered from the number of boxes received for the order under study.

Test: Select a particular purchase order for MSDs.

X5 Discrepancy in count from In boxes of MSDs by order
order
Decision: placedthe
Compute and order
value of X5 in number of boxes for the selected purchase order.
received
Criteria: Compute cycle time in days to place a shipment of MSDs in inventory by subtracting the date the shipment was received
from the date the order was place in inventory.

Test: Select a particular purchase order.

Decision: Compute the value of X6 in days for the selected purchase order.
X6 Cycle time to place product in inventory In days
Criteria: Compute inventory shelf-time in days for a box of MSDs by subtracting the date the box was placed in inventory from the
date the box was removed from inventory.

Test: Select a box of MSDs.

Decision: Compute the value of X7 in days for the selected box of MSDs.

X7 Inventory shelf time In days

Conduct Gage R&R studies on Xs.

• Recall, that the purpose of a Gage R&R study is to

determine the adequacy of the measurement system for an X.

• In this case, the measurement systems for all of the Xs are

known to be reliable and reproducible, hence, Gage R&R
studies were not conducted by team members.
Collect baseline data for each X.
• Team members gathered baseline data on durability (Y 1) and functionality
(Y2), and relevant Xs using the following sampling plan:

• For a two-week period the first box of MSDs brought to the PSD each hour
was selected as a sample; this yielded a sample of 80 boxes of MSDs

• For each sampled box, team members determined the durability (Y 1) and
functionality (Y2) measurement. Furthermore, they collected information
concerning the vendor (X1), size of the MSD (X2), whether or not the MSD
has ridges (X3), and inventory shelf life will be recorded (X 7).
• Note that Purchasing Department will separately study cycle time from order
to receipt of order (X4), discrepancy between ordered and received box counts
(X5), and cycle time from receipt of order to placement in inventory (X 6).
• These last factors may influence such concerns as choice of vendor, ordering
procedures, and inventory control, but they do not impact durability and
functionality.
• Furthermore, the MSDs are not tested after they are used so the type of
usage (X8) is not studied here.
•As was indicated in the Define phase, certain variables (e.g., X 4, X5, X6, and X8)
can be addressed in subsequent Green Belt projects.

The following show the baseline data for the Xs.

Sample Day Hour X1 X2 X3 X7 Durability Function
1 Mon 1 1 0 0 7 2 5
2 Mon 2 0 1 0 7 2 9
3 Mon 3 0 0 1 7 10 7
4 Mon 4 0 1 0 7 1 4
5 Mon 5 0 0 0 7 7 3
6 Mon 6 0 1 1 7 2 5
7 Mon 7 0 1 1 7 1 9
8 Mon 8 0 0 0 7 7 5
9 Tue 1 0 1 0 8 2 8
10 Tue 2 0 1 0 8 1 7
11 Tue 3 0 1 0 8 1 13
12 Tue 4 1 1 1 8 9 5
13 Tue 5 1 1 0 8 9 9
14 Tue 6 1 1 1 8 10 11
15 Tue 7 1 1 1 8 10 11
16 Tue 8 0 0 1 8 8 9
17 Wed 1 1 1 1 9 8 11
18 Wed 2 1 0 0 9 1 11
19 Wed 3 1 1 1 9 10 11
20 Wed 4 0 0 0 9 7 11
21 Wed 5 1 1 1 9 9 9
22 Wed 6 0 0 1 9 9 5
23 Wed 7 1 0 1 9 2 11
24 Wed 8 1 0 0 9 1 10
25 Thu 1 1 0 1 10 1 14
26 Thu 2 0 1 1 10 1 10
27 Thu 3 1 1 1 10 8 13
28 Thu 4 0 0 1 10 10 12
29 Thu 5 0 0 0 10 7 14
30 Thu 6 0 1 1 10 3 13
Sample Day Hour X1 X2 X3 X7 Durability Function
31 Thu 7 0 0 0 10 9 13
32 Thu 8 1 1 1 10 8 11
33 Fri 1 0 1 0 1 2 0
34 Fri 2 0 1 0 1 2 1
35 Fri 3 0 1 0 1 1 6
36 Fri 4 0 1 0 1 3 3
37 Fri 5 0 1 0 1 2 2
38 Fri 6 1 1 0 1 10 6
39 Fri 7 0 0 1 1 10 0
40 Fri 8 0 1 0 1 2 0
41 Mon 1 0 1 1 4 3 4
42 Mon 2 0 1 0 4 3 7
43 Mon 3 0 1 1 4 3 3
44 Mon 4 0 0 0 4 10 2
45 Mon 5 1 1 0 4 8 5
46 Mon 6 0 1 1 4 3 4
47 Mon 7 1 0 0 4 1 4
48 Mon 8 0 0 1 4 10 5
49 Tue 1 1 1 1 5 11 6
50 Tue 2 1 0 1 5 3 4
51 Tue 3 1 1 0 5 10 6
52 Tue 4 1 0 1 5 3 5
53 Tue 5 1 0 0 5 2 4
54 Tue 6 0 0 0 5 9 5
55 Tue 7 0 0 1 5 9 5
56 Tue 8 0 1 0 5 3 7
57 Wed 1 0 0 1 6 9 5
58 Wed 2 1 1 0 6 9 7
59 Wed 3 0 0 0 6 9 5
60 Wed 4 1 0 0 6 2 6
Sample Day Hour X1 X2 X3 X7 Durability Function
61 Wed 5 1 0 1 6 2 5
62 Wed 6 1 1 1 6 10 5
63 Wed 7 0 1 0 6 1 7
64 Wed 8 0 1 0 6 2 5
65 Thu 1 0 0 1 7 10 7
66 Thu 2 1 1 0 7 9 5
67 Thu 3 1 0 0 7 1 7
68 Thu 4 0 1 0 7 2 5
69 Thu 5 1 0 1 7 1 6
70 Thu 6 0 1 0 7 1 5
71 Thu 7 1 0 0 7 1 8
72 Thu 8 1 1 1 7 10 5
73 Fri 1 0 1 1 8 3 7
74 Fri 2 1 1 1 8 9 7
75 Fri 3 1 0 0 8 1 13
76 Fri 4 0 1 1 8 2 8
77 Fri 5 0 1 1 8 3 9
78 Fri 6 1 1 1 8 8 10
79 Fri 7 1 0 1 8 3 11
80 Fri 8 0 0 1 8 10 11
 
Legend:
X1 = Vendor (0 = Office Optimum and 1 = Ibix)
X2 = Size (0 = Small and 1 = Large)
X3 = Ridges (0 = Without and 1 = With)
X7 = Inventory shelf time, in days
Collect Baseline Data for Each X.
• The baseline data revealed that the yield for durability is 0.4625 (37/80)
and the yield for functionality is 0.425 (34/80), see the next table.

• As before, this indicates very poor levels for the CTQs in the Paper
Shuffling Department.

• For comparison purposes the judgment sample carried out by the team
during the Define phase showed that the yield was approximately 40%
(i.e., the team assumed the failure rate was approximately 60%) for
both durability and functionality.
• The slight increased yields in this study can be due to
natural variation in the process.

• The baseline data also showed that 56.25% of all

MSDs are from Office Optimum (X1), 42.50% of
MSDs are small (X2), 50.00% of all MSDs are without
ridges (X3), and the average shelf time for boxes of
MSDs (X7) is 6.5 days, with a standard deviation of
2.5 days, see the next table.
Basic statistics on baseline data

Variable Proportion Mean Standard

deviation

Y1: Durability 4or more 0.4625 5.213 3.703

bends/clip

Y2: Functionality 5 or fewer 0.4250 7.025 3.438

broken/box

X1: Vendor 0=Office 0.5625    

Optimum 0.4375
1=Ibix
X2: Size 0=Small 0.4250    
1=Large 0.5750

X3: Ridges 0=Without 0.5000    

Ridges 0.5000
1=With Ridges
X7: Inventory Shelf Shelf time in   6.5000 2.5160
Time days
Develop hypothesis between Xs and Ys.
• Team members develop hypotheses [Y = f (X)] about the relationships between the Xs and the Ys to identify
the Xs that are critical to improving the center, spread and shape of the Ys in respect to customer
specifications.
• This is accomplished through data mining. Data mining is a method used to analyze passive data; that is,
data that is collected as a consequence of running a process.
• In this case, the baseline data is the passive data set that will be subject to data mining procedures.
• Dot plots or box plots of durability (Y1) and functionality (Y2) stratified by X1, X2, X3 and X7 can be used to
generate some hypotheses about main effects (i.e., the individual effects of each X on Y 1 and Y2).
• Interaction plots can be used to generate hypotheses about interaction effects (i.e., those effects on Y1 or
Y2 for which the influence of one X variable depend on the level or value of another X variable) if all
combinations of levels of X variables are studied.
• If all combinations of levels of X variables are not studied, then often interaction effects are not
discovered. Frequently, screening designs are utilized to uncover interaction effects. 
• Team members constructed dot plots from the baseline data to check if any of the X’s (i.e., main effects)
impact durability (Y1) and functionality (Y2).
• The dot plots for durability an functionality are:
Dot plot for durability by X1 (i.e., vendor)
Dot plot for durability by X2 (i.e., size)
Dot plot for durability by X3 (i.e., ridges)
Dot plot for durability by X7 (i.e., shelf life)
Dot plot for functionality by X1 (i.e., vendor)
Dot plot for functionality by X2 (i.e., size)
Dot plot for functionality by X3 (i.e., ridges)
Dot plot for functionality by X7 (i.e., shelf life)
The dot plots for durability (Y1) indicate:

1. The values of durability tend to be low or high with a significant gap between 4 and 6 for X1, X2, X3 and X7, and
2. The variation in durability is about the same for all levels of X1, X2, X3 and X7.

The dot plots for functionality (Y2) indicate:

1. The values of functionality tend to be lower when X1 = 0 than when X1 = 1

2. The variation in functionality is about the same for all levels of X2 and X3
3. The values of functionality tend to be lower for low values of X7.
 
Discussion of the Analysis of Durability.

• Since there are no clear differences in variation (i.e., spread) of durability for
each of the levels of X 1, X2, X3 and X7, the team wondered if there might be
differences in the average (i.e., center) for each level of the individual X’s.

• Team members constructed a main effects plot for durability to study differences
in averages, see following figure.
Main effects plot for durability by X1, X2, X3, and X7.
Develop hypothesis between Xs and Ys.
The main effects plot for durability indicates that:

• For the range of shelf lives observed there is no clear pattern for the relationship of shelf life (X 7) to the average
Durability.
• On the other hand, it appears that ridges (i.e., X 3 = 1) has a positive relationship to the average Durability.
• At first glance it would seem that the better results for the average Durability are seen when the vendor Ibix is
chosen using small MSDs (i.e., X 1 = 1 and X2 = 0), while using large MSDs from Office Optimum
• (i.e., X1 = 0 and X2 = 1) yield worse results.
• While discussing the dot plots and main effects plot it was suggested that it is dangerous to make any
conclusions without knowing if there are interaction effects.

• An interaction effect is present when the amount of change introduced by changing one of the Xs
depends on the value of another X.

• In that case, it is misleading to choose the best value of the Xs individually without first considering the
interactions between the Xs.
• Consequently, team members did an interactions plot
for X1, X2, and X3.

• X7 was not included in the interactions plot because

the main effects plot indicated no clear pattern or
relationship with durability (Y1).

• All combinations of levels of the X variables must be

present to draw an interactions plot.

• This is often not the case with passive data (i.e., no

plan was put in place to insure all combinations were
observed in the data gathering phase).

• Fortunately, although not all combinations were

observed equally often they were all present.
Interaction effects plot for durability by X1, X2, and X3
• Interestingly, the interaction plot indicates that there is a possible interaction between X 1 (i.e., vendor) and
X2 (i.e., size).

• How is this known?

• When there is no interaction the lines should be parallel to each other indicating the amount of change in
average Durability when moving from one level of each X variable to another level should be the same for
all values of another X variable.
• This plot shows the lines on the graph of X1 and X2
not only are not parallel, but they cross.

• The average Durability is the highest when either

large Ibix MSDs (i.e., X1 = 1 and X2 = 1) or small
Office Optimum MSDs (i.e., X1 = 0 and X2 = 0) are
used.

• This means the choice of vendor may depend on the

size of MSD required.
• The main effects plot suggests that the best results for average Durability would occur when small
MSDs from Ibix are used, but the interactions plot suggests this combination would yield a bad
average Durability.

• In order to study all this further the team decides that during the Improve phase they will run a full
factorial design to examine the relationship of X1, X2, and X3 on Durability (Y1) since the main
effects plot indicates potential patterns.

• Again, there does not appear to be a relationship between Durability (Y 1) and X7.
Discussion of the Analysis of Functionality
Main effects plot for functionality by X1, X2, X3, and X7.
Interaction effects plot for functionality by X1, X2, and X3
• The main effects plot indicates that higher values of shelf life (X7) yield higher values for Functionality (Y2).

• The team surmised that the longer a box of MSDs sits in inventory (i.e., higher values of shelf life), the higher the count
of broken MSDs (i.e., functionality will be high).

• From a practical standpoint, the team felt comfortable with this conclusion.

• They decided the Purchasing Department should put a “Six Sigma” project in place to investigate whether the potential
benefit of either a “just in time” MSD ordering process or establishing better inventory handling procedures will solve the
problem.
• The interaction effects plot indicate a potential interaction between the X2 (i.e., size) and X3 (i.e., ridges).

• The better results for functionality (i.e., low values) were observed for large MSDs without ridges
(i.e., X2 = 1 and X3 = 0). Why this may be the case would need to be studied further.

• Also, there may be an interaction between X1 (i.e., vendor) and X2 (i.e., size), but it appears that better results are observed
whenever Office Optimum is used (i.e., X1 = 0).

• In other words, the average count of broken MSDs is lower (i.e., functionality average is lower) whenever Office Optimum is
the vendor.
Analyze Phase Summary.

The Analyze Phase resulted in the following hypotheses:

• Hypothesis 1: Durability = f (X1=Vendor, X2=Size, X3=Ridges) with a strong interaction effect between X1 and X2.
• Hypothesis 2: Functionality = f (X1=Vendor, X2=Size, X3 = Ridges, X7=Shelf Life) with the primary driver being X7 with
some main effect due to X1 and an interaction effect between X2 and X3.
• X7 is the main driver of the distribution of Functionality (Y2) and is under the control of the employees of POI. Hence,
team members restructured Hypothesis 2 as follows:
• Functionality = f (X1=Vendor, X2=Size, X3=Ridges) for each fixed level of X7 (Shelf Life).
 Improve Phase
The improve phase involves:

• Designing experiments to understand the relationship between the Ys and the vital few Xs and major noise variables

• Generating the actions needed to implement the levels of the vital few Xs that optimize the shape, spread and center of the distributions of the Ys

• Developing action plans

• Conducting pilot test of the action plans.

Conduct Experiments.
• First, team members conducted an experimental design to determine the effect of X1
(Vendor), X2 (Size), and X3 (Ridges), and their interactions, on the Ys with X7 = 0 (no shelf
life – MSDs are tested immediately upon arrival to POI before they are placed in inventory).

• A 23 full factorial design with 2 replications (8 trials) was performed for durability and
functionality.
• The factor conditions for Vendor (X1) are Office
Optimum (-1) or Ibix (1); the factor conditions for Size
are “Small” (-1) or “Large” (1), and the factors
conditions for Ridges (X3) are “Without Ridges” (-1)
or “With Ridges” (1).

• The experiment was setup in two blocks to increase

experimental reliability, with the first 8 runs conducted
in the morning and the second 8 runs conducted in
the afternoon.
• The runs were randomized within each block.
• The purpose of the blocks and randomization is to help prevent lurking (background) variables that are related to time
(e.g., time of day and order in which data is collected) from confusing the results.
• Additional information can be gathered since 16 trials were run, rather than the minimum of 8 trials, especially
regarding potential interactions.
• The data from the 23 full factorial design, with 2 replications in run order with the first 8 runs constituting the first
replicate, is:
Std Order Run Order Vendor Size Ridges Durability Functionality
2 1 Ibix Small Without 1 8
4 2 Ibix Large Without 9 9
3 3 Office Optimum Large Without 1 8
8 4 Ibix Large With 11 8
5 5 Office Optimum Small With 10 0
6 6 Ibix Small With 4 2
7 7 Office Optimum Large With 4 3
1 8 Office Optimum Small Without 10 2
16 9 Ibix Large With 9 3
10 10 Ibix Small Without 3 0
12 11 Ibix Large Without 9 0
14 12 Ibix Small With 3 7
13 13 Office Optimum Small With 9 6
11 14 Office Optimum Large Without 2 4
9 15 Office Optimum Small Without 8 1
15 16 Office Optimum Large With 2 4
Pareto Chart of Effects for Durability

The major effects (i.e., those that have significance level less than .
10, in other words over 90% confidence level) for Durability are the
interaction of Vendor and Size and the main effect due to Ridges.
Pareto Chart of Effects for Functionality
 There are no significant effects due to Vendor, Size,
or Ridges present for Functionality.
 This indicates that since the effect of Shelf Life was
held constant in this designed experiment, while it
was shown to affect Functionality in the data mining
analysis, the team can restrict its attention to
improving Functionality by addressing Shelf Life first.
 As Durability is the only outcome influenced by
Vendor, Size, or Ridges in this designed experiment,
further consideration in this study will be restricted to
Durability.
 Another project can address Shelf Life and its effect
on Functionality.
Since interaction effects should be interpreted prior to
studying main effects the team decided to construct
an interaction effect plot for Vendor and Size.
• The interaction effect plot between Size and Vendor indicates that the best results for
Durability are obtained using small MSDs supplied by Office Optimum or large MSDs
supplied by Ibix.

• The reasons for this interaction may be due to factors such as materials used for each
size MSD, differences in supplier processes for each size MSD, or other supplier-
dependent reasons.
• Team members can ask each vendor why their sizes
show significant differences in average durability, if
there is a preference to use only one vendor.

• Otherwise, the Purchasing Department should buy

small MSDs from Office Optimum or large MSDs
from Ibix to optimize Durability (Y1).

• The only significant main effect not involved within a

significant interaction effect is X3 = Ridges.

• The main effect for Ridges on Durability is shown

next.
• The main effect of Ridges indicates that the average Durability is about 6.5 – 5.4 = 1.1 more when an
MSD with Ridges is used rather than an MSD without Ridges.

• Therefore, since Ridges is a main effect independent of any interaction effects, then the right selection of
MSDs is to use Office Optimum for small MSDs with ridges and Ibix for large MSDs with ridges.

• If the experimental results are used, then the average Durability for Office Optimum’s small MSDs with
ridges is (10 + 9) / 2 = 9.5 while the average Durability for Ibix’s large MSDs with ridges is (11 + 9) / 2 =
10.0.
• Both averages are well above the required
corresponding CTQ of at least 4.

• As long as the variation (spread) of results is small

enough so that no individual Durability result is far
from these averages, then the team is successful
with respect to Durability.

• The variation in these results can be monitored using

control charts after changing the purchasing process
for selecting MSDs.
• The team members decided to purchase all “MSDs With Ridges.”

• In addition, the choice of Vendor and Size will be as follows: (Vendor = Office Optimum) and (Size = Small) or (Vendor =
Ibix) and (Size = Large) to maximize average durability.

• In addition, the team decided to take on another project to reduce Shelf Life to less than 5 days to address Functionality.

• The revised flowchart for the Purchasing Department incorporating the findings of the Six Sigma project is shown next.
 Start

Purchasing Department receives an order

from the Paper Shuffling Department

Purchasing agent calls Ibix to place an order A

Are large MSDs No

with ridges
available

Yes

Purchasing agent calls Office Optimum

Place an order for MSDs with Ibix

Place an order for

MSDs with Office Yes Are small MSDs
Optimum with ridges
available

Receive order from vendor

No
Store product received into inventory

Wait one day and call again

PSD removes MSDs from inventory

PSD uses MSDs

A

Stop
Pilot Test.

The team members conducted a pilot test of the

revised best practice (flowchart). Data for durability
from the pilot test is:

 
Vendor Size Ridges Durability
Hour
Shift 1 – Hour 1 Office Optimum Small With 10
Ibix Large With 11
Shift 1 – Hour 2 Office Optimum Small With 7
Ibix Large With 11
Shift 1 – Hour 3 Office Optimum Small With 10
Ibix Large With 11
Shift 1 – Hour 4 Office Optimum Small With 8
Ibix Large With 11
Shift 1 – Hour 5 Office Optimum Small With 9
Ibix Large With 10
Shift 1 – Hour 6 Office Optimum Small With 9
Ibix Large With 9
Shift 1 – Hour 7 Office Optimum Small With 8
Ibix Large With 11
Shift 1 – Hour 8 Office Optimum Small With 9
Ibix Large With 10
Shift 2 – Hour 1
Office Optimum Small With 9
Ibix Large With 11
Shift 2 – Hour 2
Office Optimum Small With 8
Ibix Large With 10
Shift 2 – Hour 3
Office Optimum Small With 10
Ibix Large With 9
Shift 2 – Hour 4
Office Optimum Small With 7
Ibix Large With 9
Shift 2 – Hour 5
Office Optimum Small With 7
Ibix Large With 10
Shift 2 – Hour 6
Office Optimum Small With 9
Ibix Large With 11
Shift 2 – Hour 7
Office Optimum Small With 10
Ibix Large With 9
Shift 2 – Hour 8
Office Optimum Small With 8
Ibix Large With 11
 
     
RTY 32/32=1
• The RTY for durability is 100%.

• Functionality was also tested, but not shown here, using Shelf Life = 0 days; that is, the MSDs were tested immediately upon
arrival to POI before they were placed in inventory.

• It resulted in a RTY of 75% that is better than the baseline RTY.

• The effect on Functionality of Shelf Life and inventory control procedures will be investigated in subsequent projects if
management decides these projects should be chartered.

• With all things being equal, large MSDs from Ibix should have a higher average durability then small MSDs from Office Optimum.
 Control Chart for Durability

Durability is in control with a higher mean number of bends for all

MSD’s in the pilot test.
Control Chart for Durability for small MSDs from Office
Optimum
Control Chart for Durability for Large MSDs from Ibix
I and MR Chart for Large MSDs f

13
UCL=12.73
12
Indiv idual Value

11
Mean=10.25
10

9
8
LCL=7.768
7
Subgroup 0 5 10 15

3 UCL=3.049
Mov ing Range

1 R=0.9333

0 LCL=0
• Durability (Y1) is in control, but it is dangerous to compute any process capability statistics
due to the small sample sizes.

• However, estimates for the mean and standard deviation of Small MSDs from Office
Optimum are 8.625 and 1.05 (calculated from the data but the calculation is not shown
here), respectively.

• The mean and standard deviation for Large MSDs from Ibix are 10.25 and 0.83,
respectively.
• Since the CTQ for durability requires the number of
bends to be 4 or more, then this requirement is 4.4
standard deviations below the mean for Small MSDs
from Office Optimum and 7.5 standard deviations
below the mean for Large MSDs from Ibix.

• Team members all agreed that as long as the

process for both Small MSDs from Office Optimum
with ridges and Large MSDs from Ibix with ridges
remain in control, then it is extremely unlikely that the
MSDs will fail the CTQ for Durability (Y1).
 The control phase involves:

• Mistake proofing the improvements and/or innovations discovered in the Six Sigma project

• Establishing a risk management plan to minimize the risk of failure of product, service or process

• Documenting the improvement and or innovation in ISO 9000 documentation

• Preparing a control plan for the process owners who will inherit the improved or innovated product,
service or process, and turn the improved process over to the process owner for continuous
improvement using the PDSA cycle
Mistake Proof the Revised Process.
Team members identified and prioritized two problems while mistake proofing the process
improvements discovered in the improve phase; they were:

1. Purchasing agents do not specify “with ridges” on a purchase order

2. Purchasing agents do not consider that the choice of vendor depends on the size of the MSDs
being requested on the purchase order.
Team members created solutions that make
both errors impossible; they are:

• The purchase-order entry system does not process

order unless “with ridges” is specified on the
purchase order
• The purchase-order entry system does not process
an order unless Office Optimum is the selected
vendor for small MSDs and Ibix is the selected
vendor for large MSDs.
Develop a Risk Management Plan.

Team members use risk management to identify two risk elements, they are:
1. Failing to train new purchasing agents in the revised purchasing process
2. Office Optimum and Ibix are out of MSDs with ridges.

Team members assigned risk ratings to both risk elements.

Risk
Risk Likelihood of Impact of element
Risk Elements Category occurrence occurrence score  

Failing to train new

purchasing agents Performance 5 5 25 High

Vendor out of MSDs

with ridges Materials 2 5 10 Medium

Scale: 1-5 with 5 being the highest.

• Both risk elements must be dealt with in risk abatement plans.

• The risk abatement plan for “failing to train new purchasing agents” is to document the revised purchasing
process in training manuals.

• The risk abatement plan for “Vendor out of MSDs with ridges” is for POI to request that both Office Optimum
and Ibix manufacture only MSDs with ridges due to their superior durability.

• This is a reasonable and acceptable suggestion to POI, Office Optimum and Ibix because the cost structures
for manufacturing MSDs with and without ridges are equal, and neither Office Optimum nor Ibix has other
customers requesting MSDs without ridges.
• Office Optimum and Ibix agree to produce only MSDs with ridges after a six-month trial period in which they check
incoming purchase orders for requests for MSDs without ridges.

• If the trial period reveals no requests for MSDs without ridges, then the POI Purchasing Department will revise the
process and the appropriate ISO 9000 documentation to reflect the possibility of purchasing only MSDs with ridges.

• Additionally, Office Optimum and Ibix thanked POI for pointing out to them that average durability is higher for MSDs
with ridges than MSDs without ridges.
• Both vendors claim that they are going to experiment
with possible different ridge patterns to increase
durability and decrease costs.

• Both vendors stated that they anticipate decrease

costs from only producing MSDs with ridges because
of the lower amortized costs of having only one
production line.
Prepare ISO Documentation.
• Team members prepare ISO 9000 documentation for
the revisions to the training manual for the purchasing
process.
Develop a Control Plan.
• This requires a monthly sampling of the boxes of
MSDs in inventory.
• The purpose of the sampling plan is to check if the
boxes of MSDs being purchased are either small
Office Optimum MSDs with ridges or large Ibix MSDs
with ridges.
• The percentage non-conforming boxes of MSDs will
be plotted on a p-chart. PSD management will use
the p-chart to highlight violations of the new and
improved purchasing process.
• The p-chart will be the basis for continuously turning
the PDSA cycle for the purchasing process.
Control Plan.
• Finally, team members checked the business indicator from the PSD
metric tracking system and determined that production costs in the PSD
decreased, probably due to the MSD Six Sigma project.

• The MSD project took effect in month 73 as per the next control chart.
 I and MR Chart for Prod Costs by C3
1 2
1400000

Individual Value 1200000

UCL=1141864
1000000
Mean=938420
800000
LCL=734977

Subgroup 0 10 20 30 40 50 60 70 80 90

1 2
400000

300000
Moving Range

UCL=249930
200000

100000
R=76495
0 LCL=0
Turn over the Process to the Process Owner

• It is time for the Champion and Black Belt to disband

the project team, turn the improved process over to the
Purchasing Department, and celebrate the project’s
success with team members.